Abstract
Quantitative assessment of cardiac function requires accurate segmentation of cardiac structures. Convolutional Neural Networks (CNNs) have achieved immense success in automatic segmentation in cardiac magnetic resonance imaging (cMRI) given sufficient training data. However, the performance of CNN models greatly degrade when the testing data is from different vendors or different centers. In this paper, we introduce the use of image registration to propagate annotation masks from labeled images to unlabeled images as to enlarge the training dataset. Furthermore, we investigated various input modalities including 3D volume, single-channel 2D image, multi-channel 2D image constructed from spatial and temporal stack to extract more features to improve domain generalization in cMRI segmentation. We evaluated our method in M&Ms-2 challenge testing data (https://www.ub.edu/mnms-2/), achieving averaged Dice scores of 0.925, 0.919 and Hausdorff Distance of 10.587 mm, 6.045 mm in right ventricular segmentation in short-axis view and long-axis view respectively.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Tran, P.V.: fully convolutional neural network for cardiac segmentation in short-axis MRI. arXiv preprint arXiv:1604.00494 (2016)
Poudel, R.P.K., Lamata, P., Montana, G.: Recurrent fully convolutional neural networks for multi-slice MRI cardiac segmentation. In: Zuluaga, M.A., Bhatia, K., Kainz, B., Moghari, M.H., Pace, D.F. (eds.) Reconstruction, segmentation, and analysis of medical images, pp. 83–94. Springer International Publishing, Cham (2017). https://doi.org/10.1007/978-3-319-52280-7_8
Glocker, B., et al.: Machine learning with multi-site imaging data: an empirical study on the impact of scanner effects. arXiv preprint arXiv:1910.04597 (2019)
Tao, Q., et al.: Deep learning-based method for fully automatic quantification of left ventricle function from cine MR images: a multivendor, multicenter study. Radiology 290(1), 81–88 (2019)
Chen, C., et al.: Improving the generalizability of convolutional neural network-based segmentation on CMR images. Front. Cardiovas. Med. 7, 105 (2020)
Zhang, Y., et al.: Semi-supervised cardiac image segmentation via label propagation and style transfer. In: Puyol Anton, E., et al. (eds.) STACOM 2020. LNCS, vol. 12592, pp. 219–227. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-68107-4_22
Sun, X., et al.: SAUN: Stack attention U-Net for left ventricle segmentation from cardiac cine magnetic resonance imaging. Med. Phys. 48(4), 1750–1763 (2021)
Isensee, F., et al.: nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat. Methods 18(2), 203–211 (2021)
Full, P.M., Isensee, F., Jäger, P.F., Maier-Hein, K.: Studying robustness of semantic segmentation under domain shift in cardiac MRI. In: Puyol Anton, E., et al. (eds.) STACOM 2020. LNCS, vol. 12592, pp. 238–249. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-68107-4_24
Campello, V.M., et al.: Multi-centre, multi-vendor and multi-disease cardiac segmentation: The M&Ms challenge. In: IEEE Transactions on Medical Imaging (2021)
Intensity-based automatic image registration - MATLAB& Simulink. https://www.mathworks.com/help/images/intensity-based-automatic-image-registration.html
Declaration
The authors of this paper declare that the segmentation methods implemented in this challenge has not used any pre-trained models nor additional MRI datasets other than those provided by the organizers.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Sun, X., Cheng, LH., van der Geest, R.J. (2022). Right Ventricle Segmentation via Registration and Multi-input Modalities in Cardiac Magnetic Resonance Imaging from Multi-disease, Multi-view and Multi-center. In: Puyol Antón, E., et al. Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge. STACOM 2021. Lecture Notes in Computer Science(), vol 13131. Springer, Cham. https://doi.org/10.1007/978-3-030-93722-5_26
Download citation
DOI: https://doi.org/10.1007/978-3-030-93722-5_26
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93721-8
Online ISBN: 978-3-030-93722-5
eBook Packages: Computer ScienceComputer Science (R0)
